Stats for Data Science
An MAA mini-course at JMM 2020, Daniel Kaplan

Day 1

Graphics
Activity 1
Printable document showing 10 "settings" for response and explanatory variables.
Data graphics chapter from Stats for Data Science book
An explanation intended for students
Example exercises Reading frames Variables in graphics Violin basics Means in context, please! Violins and n
Computing
Little App on F
A Little App where you specify response and explanatory variables and the F graphics and calculations are displayed. Invoke the Null Hypothesis by randomizing the response variable.
Point plot Little App
Displays techniques for graphing multiple variables (without statistical commentary).
Models
Revisit Activity 1
But now you're going to sketch in models of the response as a function of the explanatory variables.
The modeling chapter from Compact Inference
A very short introduction to using functions to describe relationships between response and explanatory variables.
Chapter on effect size from Compact Inference
How to compute effect size from a model
Chapter on model values from Compact Inference
These are simply the outputs of the model when the data used for fitting are given as inputs.
Example exercises Flexibility Bounded and unbounded models
Inference
Activity 2
The same 10 settings as Activity 1, but now with an additional graphic showing the distribution of model values
The measuring-variation chapter from Compact Inference
An alternative introduction to the concept of "variance" that doesn't use means.
F and R chapter from Compact Inference
Introduces R as the ratio of two variances, and F as a product of the size of R and the amount of data.
Example exercises Eyeballing variance Variance of indicator variables The two-sample t-test (using F)

Day 2

Confounding interval
A example exercise
Historical examples of confounding.
Models that learn
Achapter from the Stats for Data Science book.
Introduces recursive partitioning, a.k.a. classification and regression trees
Example exercises Learning to stratify Trees and forests
Extending models
Alas, no materials available here, yet.
Causality
Chapter from my *Statistical Modeling: A Fresh Approach*
This chapter was my attempt, circa 2011, to make Pearl's approach to causal reasoning accessible to undergraduates.
*The Book of Why* by Judea Pearl and Dana Mackenzie Reports
A compelling argument for teaming up a research scientist with a professional expositor. The historical examples in this book show some of the important ways your life has already been influenced by proper causal reasoning based on observational data.
An exercise: Education and causation.
Perhaps this exercise, which is non-trivial, will illustrate why taking causal reasoning seriously is important to informed decision making.
MAA mini-course evaluation